About this project

This site publishes the Jewish metrical books of Brzostek (Brzostek Jewish community, Galicia), covering births, deaths, and marriages from 1894 through 1938. Page-scans of the original books were transcribed and matched against a single reconciled person index. The site is a static build — every page is just HTML, JSON, and JPGs that anyone can host on a free service such as GitHub Pages, Netlify, or Cloudflare Pages.

More about Brzostek

Sources

Methodology

Every record was transcribed from the original page-scan into a spreadsheet (one workbook per book). Names were normalized for an English-speaking audience. Names fall into a two-tier system — "legal" (the form that would have been seen as legal in Vienna) and "common" (the form the family used on an everyday basis at home). A single Person Index reconciles duplicates across books: each unique person is assigned a Person_ID (P0001, P0002, …), and every record's role columns reference those IDs. Where two record entries clearly refer to the same person, the duplicate row is RETIRED (kept as a tombstone only) and merged into the canonical row. Only CONFIRMED rows are published here.

Inline name links

Inside each record's transcription, names are rendered as clickable links to the corresponding person page. The link engine is conservative: it only considers people the structured columns of that record already identify, never strangers from elsewhere in the corpus. When a name in the prose cannot be confidently linked, the small "People in this record" panel beside the transcription still names the person and links to their page; entries that the inline pass missed are tagged "linked here only".

Corrections

Each record page and each person page has a "Submit a correction" link. The link opens a new issue on the project's GitHub repository with a pre-filled template — the page reference and a blank "what's wrong / what should it say" form. A free GitHub account is required to submit. No email address is collected, and no contact information is exposed on the site. The owner reviews each submission, edits the underlying spreadsheet if appropriate, and re-runs the build to apply the correction.

Privacy

This site has no analytics, no cookies, and no third-party trackers. The only external resources it loads are MiniSearch and Chart.js from public CDNs — both static JavaScript libraries, both delivered without user data. Corrections are routed through GitHub Issues; submission content and the submitter's GitHub username become part of the public issue thread.

Credits

Translation, transcription, and reconciliation: project team. Page-scans contributed by descendants and researchers. Site build by the project's static-site pipeline.